A Performance Margin and Reliability Analysis Method for Fatigue Performance of Welded Thin-walled Pipe Structures

Traditional probability-based methods in welding can achieve reliability assessment but fail to establish a direct and actionable connection between design parameters and reliability, which significantly hinders the forward design process. Meanwhile, performance margin-based assessment methods lack established universal standards and norms for reliability evaluation and allocation, thereby impeding their effective implementation in engineering practice. This study aims to rigorously establish a conversion relationship between performance margins and reliability, enabling seamless integration with traditional probabilistic reliability systems. Based on the Quantification of Margins and Uncertainties (QMU) framework, we propose an innovative fatigue life assessment method for thin-walled welded tube structures that comprehensively considers multi-source uncertainties. This method accurately quantifies the impact of various welding fatigue-related random variables on fatigue life. Additionally, a robust correlation model between confidence factors and reliability is established, successfully obtaining the mapping relationship among welding fatigue life, performance margins, and reliability. This enables the seamless conversion and correlation between margin-based reliability assessment and traditional reliability assessment, demonstrating significant engineering value.